The present invention relates to novel antithrombotic peptides and, more particularly, to short peptides that mediate divalent cation-independent adhesion of platelets to fibronectin. These peptides recognize and/or inhibit adhesion in the first type II (type II-1) repeat element of fibronectin. This repeat element exists in the 45 kDa gelatin-binding domain of fibronectin.
NOTE: Literature references on the following background information and on conventional test methods and laboratory procedures well-known to the ordinary person skilled in the art, and other such state-of-the-art techniques as used herein, are indicated by reference numbers in parentheses, and appended at the end of the specification.
Fibronectin (FN) is a large glycoprotein which is found in plasma and extracellular matrices and is associated with cell membranes where it has been shown to play an important role in mediating the adhesive properties of platelets and other cell types (for reviews, see references 1 and 2). The FN molecule is composed of a series of repeat elements, which are grouped linearly along the polypeptide chain to form functional domains (see FIG. 1). Several of these domains have been implicated in the adhesive activities of FN.
The pioneering studies of Pierschbacher and Ruoslahti (3) established the critical role of the RGD sequence in mediating the adhesion-promoting properties of the cell-binding domain. More recent studies have indicated that adjacent flanking sequences also contribute to the adhesive properties within this domain (4,5,6). At least two further sequences within the alternately-spliced IIICS repeat (LDV and REDV) may also help mediate adhesive activity for some cell types (7,8).
Furthermore, several studies ((9,10), for example) have implicated at least two sites within the high affinity carboxy terminal heparin-binding domain in promoting cell adhesion to FN, while a site within the amino terminal 27 kDa of FN is thought to interact with the matrix assembly receptor (11,12). Clearly the interactions of cells with FN are complex and may be mediated by a variety of receptors interacting with distinct sites on the FN molecule.
An early response to vascular injury is the deposition of platelets on the exposed subendothelial surface. Under the conditions of flow, this deposition has been shown to depend upon the presence of both FN and von Willebrand Factor (13,14,15). In simpler model systems, FN has been shown to be necessary for platelet deposition onto collagenous substrates (16), and under static conditions FN has been shown to mediate platelet spreading on collagen substrates (17,18).
To date only two FN receptors have been identified on platelet surfaces, the .alpha..sub.5 .beta..sub.1 integrin (GP Ie-IIa) complexes (19,20). As integrins, both of these receptors require the presence of divalent cations for ligand binding activity (21). In addition, while the .alpha..sub.IIb -.beta..sub.3 complex binds FN, fibrinogen, vitronectin and Von Willebrand Factor, it requires activation to do so (20).
Nievelstein and Sixma (14) studied the FN-dependent adhesion of platelets to collagen-coated surfaces and to the extracellular matrix of cultured endothelial cells under conditions of flow. On the basis of experiments with RGD-containing peptides, which inhibit both .alpha..sub.5 .beta..sub.1 and .alpha..sub.IIb .beta..sub.3 function, platelets from patients with Glanzmann's thrombasthenia, which lack .alpha..sub.IIb .beta..sub.3, and inhibitory anti-bodies directed against the .alpha..sub.IIb .beta..sub.3 and .alpha..sub.5 .beta..sub.1 integrins, they concluded that there must exist yet another "binding system" on platelets for FN which is operational when FN is adsorbed onto a surface.
Fibronectin has been shown to in fact undergo a conformational change upon adsorption onto surfaces and this change results in exposure of initially cryptic determinants (22,23). Previous studies by the present inventor and his group and by Piotrowicz et al (19, 24) indicated that 20-50% of total platelet adhesion to FN in a static system was resistant to inhibition by either EDTA or by a combination of saturating concentrations of antibodies directed against the .alpha..sub.IIb .beta..sub.3 and .alpha..sub.5 .beta..sub.1 integrins, further substantiating the presence of an additional adhesive mechanism.
Recently, Winters et al (25) demonstrated that the divalent cation-independent adhesion of platelets to FN is mediated by the 45 kDa gelatin-binding domain, which is released from FN by digestion with thermolysin. Unlike the .alpha..sub.5 .beta..sub.1 integrin-mediated-, divalent cation-dependent platelet adhesion which leads to extensive platelet spreading, this gelatin-binding domain supported-, divalent cation-independent platelet adhesion does not support the spreading of platelets on FN-coated substrates.
Furthermore, this divalent cation-independent adhesion is resistant to either reduction and alkylation nor by enzymatic deglycosylation of the 45 kDa domain. Neither peptides containing the RGD- or the DGEA-sequences nor inhibitory antibodies directed against the .alpha..sub.IIb .beta..sub.3, .alpha..sub.v .beta..sub.3, .alpha..sub.2 .beta..sub.1 and .alpha..sub.5 .beta..sub.1 integrins, the GPIb-IX complex (Von Willebrand factor receptor) or against thrombospondin inhibited the divalent cation-independent adhesive mechanism.
These results indicate that platelets have an adhesive system, which permits them to adhere to the gelatin-binding domain of FN in a divalent cation-independent manner, and that the recognition site for the receptor(s) mediating this mechanism appears to consist of a linear polypeptide sequence within the 45 kDa gelatin-binding domain of FN. Accordingly, it is an object of this invention to determine the minimal sequence of amino acids constituting the adhesion recognition site for this mechanism.